Abstract 3641: Hemoglobin is a Predictor of All-Cause Mortality and Cardiovascular Mortality in Patients Undergoing Exercise Treadmill Testing
Background: While hemoglobin is a known predictor of cardiovascular (CV) death following acute myocardial infarction, little is known about the role of hemoglobin level in predicting all-cause and CV mortality in patients referred for routine exercise testing. We hypothesized that subjects with low hemoglobin would have an increase risk of all-cause and CV mortality.
Methods: Clinical variables, laboratory values, treadmill test and event data were obtained for 1,959 subjects that underwent exercise treadmill testing for routine clinical indications between 1996 to 2003. All-cause mortality was obtained from the social security death index and autopsy reports or clinical notes were used to determine CV mortality. Student T-test and Chi-square analysis were used to compare continuous and dichotomous variables between patients that lived and died during follow-up. Cox proportional hazards analysis and Kaplan-Meier survival analysis were performed to determine which variables predicted all-cause and CV death. P-values<0.05 were considered significant.
Results: The mean age of our population was 57±12 years, 95% were male, 16% had diabetes, 51% had hypertension, 35% had hypercholesterolemia, and 66% had a history of smoking. During follow-up, 10% of subjects died, 3.8% of subjects died of CV causes, and 10.6% had a cardiovascular event defined as CV death, MI, and ACS. In male patients that died, hemoglobin was significantly lower than patients that did not die(14.2±0.11 vs 14.8±0.03, p<0.0001). Cox proportional hazard analysis demonstrated that age, known coronary artery disease, hemoglobin, and METs were significant predictors of CV mortality. Age, heart rate recovery at two minutes, hemoglobin, and METs were significant predictors of all-cause mortality. Kaplan-Meiers survival analysis demonstrated that patients with anemia (Hgb<13) had significantly higher mortality than patients with hemoglobin<13 (Odds Ratio 2.2, 95% CI 1.4 to 3.4, p<0.0005).
Conclusion: Hemoglobin is a powerful predictor of mortality and provides additional prognostic information to that available from exercise stress testing.